If you want to do this from within SAS, XCMD has to be enabled.
Note that non-root users can only kill processes they own; if you need to kill other processes, you need superuser (root) privileges.
... View more
Use a DATA step (simpler code).
Define a maximum length (32767) character variable.
Proceed through the macro variable (COUNTW and SCAN functions), use the QUOTE function to add the single quotes, and CATX to store the results in the new variable.
CALL SYMPUTX to create the new macro variable.
... View more
If it's not always there, then it's of no use for proper analysis, which also means there's no reason to explicitly KEEP it.
If, OTOH, you really need it, then you will have to infer values for the missing periods, which also means dynamic code where you omit the KEEP and therefore do not need to suppress the WARNING/ERROR.
... View more
This is why we advocate against having macro code in step (DATA or PROC) code.
Had you put the %LET before the DATA step (illustrating the actual sequence of execution), the mistake would have been obvious. Had you put it after the RUN, it would simply have worked as intended.
BTW you overcomplicate it. If the goal is to get cyear, this is sufficient:
data _null_;
infile oscmd;
input;
call symputx("cyear",substr(_infile_,65,4));
run;
... View more
For which purpose do you need that number? It cannot be used in further analysis, and if you need to export data in this format, the YYMMDDN format can handle this without conversion.
... View more
And another thing: if there's just one ID with lots of observations, the resulting dataset will be enormous and consist mainly of missing values.
... View more
If people need to look at it, it's a report, not a dataset. Hand them the PROC REPORT code so they can run it on the intended subset (nobody will manually inspect 17 million observations in a single lifetime).
... View more
I strongly encourage you to read Using the Hash Object to Store and Retrieve Data and come back with any further questions you have that need clarification.
... View more
I merged your identical questions; posting a question once is sufficient. Superusers can move a topic to a different community if this is necessary.
... View more
Maxim 1: Read the Documentation
METADATA_RESOLVE Function :
rc = METADATA_RESOLVE(uri, type, ID);
The function needs three arguments, you only supply two.
... View more
A simple process like writing a dataset to an external text file is ALWAYS I/O bound. Given your further description, the main bottleneck is the download from the remote server (database or base SAS library on a network share, or anything else? Please clarify), followed by the bandwidth of the local storage where WORK resides.
If you fetch from a database, the observed fluctuations may come from network saturation and/or load on the database system itself.
... View more
Look for ODS statements rerouting your output to somewhere else (a pdf or xlsx file, for instance). This might have happened in previous steps/submits.
... View more
Assign a $HEX format of suitable length (double the defined length of the variable) and look at the hex codes. This will give you an idea of what is really stored there (leading blanks, possible UTF characters, non-printable characters).
... View more
